Inhibitory Effects of Cytosolic Ca2+ Concentration by Ginsenoside Ro Are Dependent on Phosphorylation of IP3RI and Dephosphorylation of ERK in Human Platelets

Intracellular Ca2+ ([Ca2+]i) is platelet aggregation-inducing molecule and is involved in activation of aggregation associated molecules. This study was carried out to understand the Ca2+-antagonistic effect of ginsenoside Ro (G-Ro), an oleanane-type saponin in Panax ginseng. G-Ro, without affecting leakage of lactate dehydrogenase, dose-dependently inhibited thrombin-induced platelet aggregation, and the half maximal inhibitory concentration was approximately 155 μM. G-Ro inhibited strongly thrombin-elevated [Ca2+]i, which was strongly increased by A-kinase inhibitor Rp-8-Br-cAMPS compared to G-kinase inhibitor Rp-8-Br-cGMPS. G-Ro increased the level of cAMP and subsequently elevated the phosphorylation of inositol 1, 4, 5-triphosphate receptor I (IP3RI) (Ser1756) to inhibit [Ca2+]i mobilization in thrombin-induced platelet aggregation. Phosphorylation of IP3RI (Ser1756) by G-Ro was decreased by PKA inhibitor Rp-8-Br-cAMPS. In addition, G-Ro inhibited thrombin-induced phosphorylation of ERK 2 (42 kDa), indicating inhibition of Ca2+ influx across plasma membrane. We demonstrate that G-Ro upregulates cAMP-dependent IP3RI (Ser1756) phosphorylation and downregulates phosphorylation of ERK 2 (42 kDa) to decrease thrombin-elevated [Ca2+]i, which contributes to inhibition of ATP and serotonin release, and p-selectin expression. These results indicate that G-Ro in Panax ginseng is a beneficial novel Ca2+-antagonistic compound and may prevent platelet aggregation-mediated thrombotic disease.

cAMP-and cGMP-dependent protein kinase (PKA and PKG), which phosphorylates substrate protein, IP 3 RI. The action of IP 3 RI is inhibited by its phosphorylation, and IP 3 RI phosphorylation is involved in inhibition of [Ca 2+ ] mobilization [10][11][12]. Therefore, phosphorylating IP 3 RI is very useful for evaluating the Ca 2+ -antagonistic effect of substances or compounds. With regard to the influx of extracellular Ca 2+ across PM, it is well known that phosphatidylinositol 3kinase and phosphatidylinositol 4-kinase (PI3K and PI4K) and extracellular signal-regulated kinase (ERK) are involved in the entry of extracellular Ca 2+ across PM by activating the Ca 2+ -permeable channel protein and human transient receptor potential channel (hTrp1) coupling with IP 3 R type II (IP 3 RII) [13][14][15][16][17][18]. Ginseng, the root of Panax ginseng Meyer, has been used frequently in traditional oriental medicine and is known to have various pharmacological activities such as anti-inflammatory action, antioxidation, antitumor, antidiabetes, and antihepatotoxicity [21,22]. In recent, it is reported that Korean red ginseng has an effect on cardiovascular disease, which is characterized with regard to reduction of blood pressure and arterial stiffness by inhibition of Rho kinase [23], anticoagulation by prolonged prothrombin and activated partial thromboplastin time [24], endothelium relaxation by nitric oxide-cGMP pathway [25], and inhibition of hypercholesterolemia-induced platelet aggregation [26]. In our previous report, we demonstrated that total saponin from Korean red ginseng (TSKRG) is a beneficial traditional oriental medicine in platelet-mediated thrombotic disease via suppression of cyclooxygenase-1 (COX-1) and thromboxane A 2 synthase (TXAS) to inhibit production of thromboxane A 2 (TXA 2 ) [27]. It is reported that ginsenoside Ro (G-Ro, Figure 1) has no inhibitory effect on collagen-elevated [Ca 2+ ] [28]. However, in this study, we showed that G-Ro attenuates thrombin-elevated [Ca 2+ ] via cAMP-dependent IP 3 RI phosphorylation and dephosphorylation of ERK in human platelets. This study provides novel information for antiplatelet effects of G-Ro in ginseng.

Preparation of Washed Human Platelets.
Human plateletrich plasma (PRP) anticoagulated with acid-citrate-dextrose solution (0.8% citric acid, 2.2% sodium citrate, and 2.45% glucose) was obtained from Korean Red Cross Blood Center (Changwon, Korea). PRP was centrifuged for 10 min at 125 g to remove a few red blood cells and white blood cells and was centrifuged for 10 min at 1,300 g to obtain the platelet pellets. The platelets were washed twice with washing buffer (138 mM NaCl, 2.7 mM KCl, 12 mM NaHCO 3 , 0.36 mM NaH 2 PO 4 , 5.5 mM glucose, and 1 mM Na 2 EDTA, pH 6.5).
Evidence-Based Complementary and Alternative Medicine 3 The washed platelets were then resuspended in suspension buffer (138 mM NaCl, 2.7 mM KCl, 12 mM NaHCO 3 , 0.36 mM NaH 2 PO 4 , 0.49 mM MgCl 2 , 5.5 mM glucose, and 0.25% gelatin, pH 6.9) to a final concentration of 5 × 10 8 /mL. All of the above procedures were carried out at 25 ∘ C to avoid platelet aggregation from any effect of low temperature. The Korea National Institute for Bioethics Policy Public Institutional Review Board (Seoul, Korea) approved these experiments (PIRB12-072).

Measurement of Platelet
Aggregation. Human washed platelets (10 8 /mL) were preincubated for 3 min at 37 ∘ C in the presence of 2 mM exogenous CaCl 2 with or without substances and then stimulated with thrombin (0.05 U/mL) for 5 min. Aggregation was monitored using an aggregometer (Chrono-Log, Corporation) at a constant stirring speed of 1,000 rpm. Each aggregation rate was calculated as an increase in light transmission. The suspension buffer was used as the reference (transmission 0). G-Ro was dissolved in platelet suspension buffer (pH 6.9).

Lactate Dehydrogenase Activity Assay.
Human platelet cytotoxicity was determined by the leakage of lactate dehydrogenase (LDH) from cytosol. Human washed platelets (10 8 /mL) were incubated for 5 min at 37 ∘ C with various concentrations of G-Ro and then centrifuged at room temperature for 2 min at 12,000 g. The supernatant was measured by LDH assay kit (Cayman Chemical) at an optical density of 490 nm. LDH leakage is expressed as the percentage of the total enzyme activity in platelets completely lysed with 0.1% Triton X-100.

Determination of Cytosolic Free Ca 2+ ([Ca 2+ ] )
. PRP was incubated with 5 M Fura 2-AM at 37 ∘ C for 60 min. Because Fura 2-AM is light sensitive, the tube containing the PRP was covered with aluminum foil during loading. The Fura 2loaded washed platelets were prepared using the procedure described above and 10 8 platelets/mL were preincubated for 3 min at 37 ∘ C with or without G-Ro in the presence of 2 mM CaCl 2 and then stimulated with thrombin (0.05 U/mL) for 5 min for evaluation of [Ca 2+ ] . Fura 2 fluorescence was measured with a spectrofluorometer (SFM 25, BioTeck Instrument, Italy) with an excitation wavelength that was changed every 0.5 sec from 340 to 380 nm; the emission wavelength was set at 510 nm. The [Ca 2+ ] values were calculated using the method of Grynkiewicz [29].

Determination of ATP and Serotonin Release.
Washed platelets (10 8 /mL) were preincubated for 3 min at 37 ∘ C with or without G-Ro and other reagents in the presence of 2 mM CaCl 2 and then stimulated with thrombin (0.05 U/mL). The reaction was terminated and centrifuged with 200 g at 4 ∘ C for 10 min, and supernatants were used for the assay of ATP and serotonin release. ATP release was measured in a luminometer (BioTek Instruments) using an ATP assay kit. Serotonin release was measured with a Synergy HT Multi-Model Microplate Reader (BioTek Instruments, Winooski, VT, USA) using serotonin ELISA kit.

Determination of p-Selectin Release.
Washed human platelets (10 8 /mL) were preincubated for 3 min at 37 ∘ C with or without substances in the presence of 2 mM CaCl 2 and then stimulated with thrombin (0.05 U/mL). The platelets were reconstituted by ice-cold phosphate-buffered saline (PBS, pH 7.4) 250 L and cells were incubated with Alexa Fluor 488 anti-human CD62P (10 L) in PBS (pH 7.4), containing 0.09% sodium azide and 0.2% bovine serum albumin (BSA) for 60 min at 4 ∘ C in the dark room. Next, platelets were washed three times by ice-cold PBS and resuspended by 0.5% paraformaldehyde in PBS. Alexa Fluor 488 anti-human CD62P binding to platelets was determined using flow cytometry (BD Biosciences, San Diego, CA, USA) and data were analyzed using CellQuest software. differences among the group means, then each group was compared by the Newman-Keuls method. Statistical analysis was performed according to the SPSS 21.0.0 (SPSS, Chicago, IL, USA). < 0.05 was considered to be statistically significant.

Effects of Cytotoxicity of G-Ro to Human Platelets.
Cytotoxicity of drugs is evaluated by cytosolic LDH leakage, which is different from platelet aggregation or granule secretion by platelet agonists [30][31][32]. Therefore, we investigate the effect of G-Ro on cytotoxicity to human platelets. When human platelets (10 8 /mL) were treated by membrane detergent Triton X-100 as a positive control, LDH was potently released ( Figure 3). However, G-Ro (50 to 300 M) that inhibited thrombin-induced platelet aggregation did not release LDH as compared to that by Triton X-100. LDH leakage by various concentrations of G-Ro (50, 100, 200, and 300 M) was 2.1% (at 50 M of G-Ro), 2.2% (at 100 M of G-Ro), 2.5% (at 200 M of G-Ro), and 2.9% (at 300 M of G-Ro), respectively, which were not significantly different from that (1.8%) by resting platelets (Figure 3).  Triton X-100 Figure 3: Effects of G-Ro on cytotoxicity. Measurement of cytotoxicity was carried out as described in "Section 2." For a positive control, 0.1% Triton X-100 was used to treat platelets. The data are expressed as the mean ± standard deviation ( = 4). NS, not significant versus without G-Ro, control.

Effects of G-Ro on cAMP and cGMP Production.
Because it was confirmed that inhibition of [Ca 2+ ] level by G-Ro is dependent on cAMP/PKA and cGMP/PKG, next, we investigated whether G-Ro increases cAMP and cGMP production in thrombin-induced platelet aggregation. As shown in Figure 5(a), thrombin weakly decreased cAMP level, but G-Ro (50 to 300 M) dose-dependently increased thrombinattenuated cAMP level. With regard to cGMP, thrombin did not change cGMP level as compared with that of basal level; on the contrary, cGMP appears to be decreased by G-Ro ( Figure 5(b)).

Effects of G-Ro on Dephosphorylation of ERK.
It is known that Ca 2+ mobilized from DTS (ER) is involved in phosphorylation of PI3K and ERK to influx extracellular Ca 2+ [13,14,17,18]. Here, we investigated the effect of G-Ro on dephosphorylation of ERK (1/2). As shown in Figure 7 inhibited thrombin-induced phosphorylation of ERK 2 (42 kDa), and the ratio of p-ERK 2 to ERK 2 (42 kDa) (Figure 7, Lanes 3 to 6).

Effects of G-Ro on ATP and Serotonin Release from Dense
Body. Dense body contains nucleotides (ATP and ADP) and serotonin; these are released by collagen-or thrombinelevated [Ca 2+ ] and subsequently involve amplification of platelet activation [33][34][35][36]. Because G-Ro decreased thrombin-elevated [Ca 2+ ] level, we investigated whether G-Ro is involved in inhibition of ATP and serotonin release. G-Ro (50 to 300 M) dose-dependently inhibited thrombin-induced ATP (Figure 8(a)) and serotonin release (Figure 8(b)). Next, we investigated whether the inhibition of ATP and serotonin release by G-Ro was dependent on cAMP/PKA-pathway and cGMP/PKG-pathway. G-Ro-downregulated (300 M) release of ATP and serotonin was significantly increased in the presence of PKA inhibitor Rp-8-Br-cAMPS and PKG inhibitor Rp-8-Br-cGMPS (Figure 8(c)). In addition, these inhibitors significantly elevated G-Ro-decreased (155, 300 M) serotonin release (Figure 8(d)).

Effects of G-Ro on p-Selectin Expression.
Compounds in -granule of platelets are known to be involved in inflammation, coagulation, and angiogenesis [36]; these are also Ca 2+dependently released by various platelet agonists. In particular, p-selectin is released from -granule and is reexpressed to the platelet surface [37] and subsequently is involved in inflammation by binding to the p-selectin glycoprotein ligand-1 receptor on monocyte [38]. Therefore, we investigated the effect of G-Ro on p-selectin expression fromgranule. Thrombin stimulated the expression of p-selectin (Figure 9(a)-(B)) as compared with that by unstimulated platelets (Figure 9

Discussion
Thrombin-elevated [Ca 2+ ] is involved in various cellular events to activate platelets, which leads to granule secretion and platelet aggregation. IP 3 that is generated from PIP 2 by agonists (i.e., thrombin, collagen, and ADP) mobilizes [Ca 2+ ] from Ca 2+ store DTS via IP 3 RI. Depletion of the intracellular Ca 2+ store by IP 3 is known to connect the influx of extracellular Ca 2+ , which is stimulated by PI3K and ERK [14][15][16][17][18]. Even though G-Ro inhibited thrombin-elevated [Ca 2+ ] , it is not unknown if G-Ro inhibited the production of IP 3 via inhibition of phospholipase C . The Ca 2+ -antagonistic reaction by cAMP and cGMP is mediated by PKA/IP 3 RI and PKG/IP 3 RI phosphorylation pathways, respectively. Even if G-Ro would not inhibit PLC-activity and IP 3 production in   (20 kDa), which is involved in granule secretion such as ATP and serotonin [8,9] to intensify platelet aggregation. It is known that G-Ro does not reduce collagen-elevated [Ca 2+ ] level [40]; however, in this study, we confirmed that G-Ro decreases thrombin-elevated [Ca 2+ ] level by stimulating the cAMP-dependent phosphorylation of IP 3 RI (Ser 1756 ) and dephosphorylation of ERK 2 (42 kDa). Therefore, it is thought that the inhibition of ATP and serotonin secretion by G-Ro resulted from the inhibition of IP 3 RI-mediated [Ca 2+ ] mobilization, and ERK 2-mediated (42 kDa) Ca 2+ influx.
Platelet aggregation is generated at site of vascular wall injury and is involved in the formation of thrombus. During the formation of thrombus, platelets release cell growth proteins such as platelet-derived growth factor (PDGF) and vascular endothelial growth factor (VEGF) in -granule [41,42]. It is well established that PDGF and VEGF induce the proliferation of fibroblast, vascular smooth cells, and epithelial cells and subsequently enhance the rate of atherosclerosis lesion progression [43][44][45][46][47]. The progression of atherosclerosis is strongly induced by inflammatory cell such as monocyte/macrophage and neutrophil [48]. Although G-Ro has antiplatelet effects, if G-Ro does not inhibit inflammation by leukocyte, the progression of atherosclerosis lesion would be generated at site of vascular wall injury, and a question for antiplatelet effects of G-Ro might be raised. ATP and serotonin released from dense body are known to be involved in amplification of platelet aggregation [33][34][35][36], and pselectin released from -granule is known to be involved in causing of inflammation [36][37][38]. Because G-Ro inhibited the release of ATP, serotonin, and p-selectin, it is thought that G-Ro may inhibit aggregation-amplification and inflammation. In real, G-Ro is known to have anti-inflammatory activity in vivo and in vitro [49,50]. It is well reviewed that ginsenosides have anti-inflammatory effects by inhibiting the production of various proinflammatory mediators (i.e., PGE 2 and NO) Evidence-Based Complementary and Alternative Medicine  . Determination of p-selectin expression was carried out as described in "Section 2." The data are expressed as the mean ± standard deviation ( = 4). * < 0.05 versus the thrombin-stimulated human platelets; † < 0.05 versus the thrombin-stimulated human platelets in the presence of G-Ro (300 M). [51]. With regard to the composition of G-Ro, Sanada et al. [19] reported that G-Ro is contained in Panax ginseng, and Choi [20] very well reviewed that G-Ro (0.045 w/w %) is contained in Panax ginseng, but not in Panax notoginseng (Sanchi ginseng). It has been reported that G-Ro inhibited arachidonic acid-induced platelet aggregation and fibrin formation in vitro [52] and activated fibrinolysis, indicating the inhibition of fibrin thrombi ex vivo [53]. Considering these previous reports [49][50][51][52][53], it is thought that G-Ro may have antithrombotic and antiatherosclerotic effects without generation of inflammation and progression of atherosclerotic lesion at site of vascular wall injury. Therefore, G-Ro is highlighted as a nontoxic antiplatelet compound together with effect that G-Ro did not affect LDH leakage.
In conclusion, the most important result of this study is that G-Ro cAMP-dependently phosphorylates IP 3 RI (Ser 1756 ) and dephosphorylates ERK 2 (42 kDa) to reduce thrombinelevated [Ca 2+ ] level, which contributed to attenuating the release of ATP, serotonin, and p-selectin. In addition, G-Ro could be clinically applied to the prevention of plateletmediated thrombosis. Therefore, it is thought that G-Ro may represent a useful tool in the therapy and prevention of vascular diseases associated with platelet aggregation.